diff src/testdir/test_regexp_utf8.vim @ 18906:bb87c5c1e29b v8.2.0014

patch 8.2.0014: test69 and test95 are old style Commit: https://github.com/vim/vim/commit/afc13bd8271819c7871ff2ae2cfebb22190a0d39 Author: Bram Moolenaar <Bram@vim.org> Date: Mon Dec 16 22:43:31 2019 +0100 patch 8.2.0014: test69 and test95 are old style Problem: Test69 and test95 are old style. Solution: Convert to new style tests. (Yegappan Lakshmanan, closes https://github.com/vim/vim/issues/5365)
author Bram Moolenaar <Bram@vim.org>
date Mon, 16 Dec 2019 22:45:04 +0100
parents f4ce361bb1e5
children 097f5b5c907b
line wrap: on
line diff
--- a/src/testdir/test_regexp_utf8.vim
+++ b/src/testdir/test_regexp_utf8.vim
@@ -215,3 +215,144 @@ func Test_optmatch_toolong()
   set re=0
 endfunc
 
+" Test for regexp patterns with multi-byte support, using utf-8.
+func Test_multibyte_chars()
+  " tl is a List of Lists with:
+  "    2: test auto/old/new  0: test auto/old  1: test auto/new
+  "    regexp pattern
+  "    text to test the pattern on
+  "    expected match (optional)
+  "    expected submatch 1 (optional)
+  "    expected submatch 2 (optional)
+  "    etc.
+  "  When there is no match use only the first two items.
+  let tl = []
+
+  " Multi-byte character tests. These will fail unless vim is compiled
+  " with Multibyte (FEAT_MBYTE) or BIG/HUGE features.
+  call add(tl, [2, '[[:alpha:][=a=]]\+', '879 aiaãâaiuvna ', 'aiaãâaiuvna'])
+  call add(tl, [2, '[[=a=]]\+', 'ddaãâbcd', 'aãâ'])								" equivalence classes
+  call add(tl, [2, '[^ม ]\+', 'มม oijasoifjos ifjoisj f osij j มมมมม abcd', 'oijasoifjos'])
+  call add(tl, [2, ' [^ ]\+', 'start มabcdม ', ' มabcdม'])
+  call add(tl, [2, '[ม[:alpha:][=a=]]\+', '879 aiaãมâมaiuvna ', 'aiaãมâมaiuvna'])
+
+  " this is not a normal "i" but 0xec
+  call add(tl, [2, '\p\+', 'ìa', 'ìa'])
+  call add(tl, [2, '\p*', 'aあ', 'aあ'])
+
+  " Test recognition of some character classes
+  call add(tl, [2, '\i\+', '&*¨xx ', 'xx'])
+  call add(tl, [2, '\f\+', '&*Ÿfname ', 'fname'])
+
+  " Test composing character matching
+  call add(tl, [2, '.ม', 'xม่x yมy', 'yม'])
+  call add(tl, [2, '.ม่', 'xม่x yมy', 'xม่'])
+  call add(tl, [2, "\u05b9", " x\u05b9 ", "x\u05b9"])
+  call add(tl, [2, ".\u05b9", " x\u05b9 ", "x\u05b9"])
+  call add(tl, [2, "\u05b9\u05bb", " x\u05b9\u05bb ", "x\u05b9\u05bb"])
+  call add(tl, [2, ".\u05b9\u05bb", " x\u05b9\u05bb ", "x\u05b9\u05bb"])
+  call add(tl, [2, "\u05bb\u05b9", " x\u05b9\u05bb ", "x\u05b9\u05bb"])
+  call add(tl, [2, ".\u05bb\u05b9", " x\u05b9\u05bb ", "x\u05b9\u05bb"])
+  call add(tl, [2, "\u05b9", " y\u05bb x\u05b9 ", "x\u05b9"])
+  call add(tl, [2, ".\u05b9", " y\u05bb x\u05b9 ", "x\u05b9"])
+  call add(tl, [2, "\u05b9", " y\u05bb\u05b9 x\u05b9 ", "y\u05bb\u05b9"])
+  call add(tl, [2, ".\u05b9", " y\u05bb\u05b9 x\u05b9 ", "y\u05bb\u05b9"])
+  call add(tl, [1, "\u05b9\u05bb", " y\u05b9 x\u05b9\u05bb ", "x\u05b9\u05bb"])
+  call add(tl, [2, ".\u05b9\u05bb", " y\u05bb x\u05b9\u05bb ", "x\u05b9\u05bb"])
+  call add(tl, [2, "a", "ca\u0300t"])
+  call add(tl, [2, "ca", "ca\u0300t"])
+  call add(tl, [2, "a\u0300", "ca\u0300t", "a\u0300"])
+  call add(tl, [2, 'a\%C', "ca\u0300t", "a\u0300"])
+  call add(tl, [2, 'ca\%C', "ca\u0300t", "ca\u0300"])
+  call add(tl, [2, 'ca\%Ct', "ca\u0300t", "ca\u0300t"])
+
+  " Test \Z
+  call add(tl, [2, 'ú\Z', 'x'])
+  call add(tl, [2, 'יהוה\Z', 'יהוה', 'יהוה'])
+  call add(tl, [2, 'יְהוָה\Z', 'יהוה', 'יהוה'])
+  call add(tl, [2, 'יהוה\Z', 'יְהוָה', 'יְהוָה'])
+  call add(tl, [2, 'יְהוָה\Z', 'יְהוָה', 'יְהוָה'])
+  call add(tl, [2, 'יְ\Z', 'וְיַ', 'יַ'])
+  call add(tl, [2, "ק\u200d\u05b9x\\Z", "xק\u200d\u05b9xy", "ק\u200d\u05b9x"])
+  call add(tl, [2, "ק\u200d\u05b9x\\Z", "xק\u200dxy", "ק\u200dx"])
+  call add(tl, [2, "ק\u200dx\\Z", "xק\u200d\u05b9xy", "ק\u200d\u05b9x"])
+  call add(tl, [2, "ק\u200dx\\Z", "xק\u200dxy", "ק\u200dx"])
+  call add(tl, [2, "\u05b9\\Z", "xyz"])
+  call add(tl, [2, "\\Z\u05b9", "xyz"])
+  call add(tl, [2, "\u05b9\\Z", "xy\u05b9z", "y\u05b9"])
+  call add(tl, [2, "\\Z\u05b9", "xy\u05b9z", "y\u05b9"])
+  call add(tl, [1, "\u05b9\\+\\Z", "xy\u05b9z\u05b9 ", "y\u05b9z\u05b9"])
+  call add(tl, [1, "\\Z\u05b9\\+", "xy\u05b9z\u05b9 ", "y\u05b9z\u05b9"])
+
+  " Combining different tests and features
+  call add(tl, [2, '[^[=a=]]\+', 'ddaãâbcd', 'dd'])
+
+  " Run the tests
+  for t in tl
+    let re = t[0]
+    let pat = t[1]
+    let text = t[2]
+    let matchidx = 3
+    for engine in [0, 1, 2]
+      if engine == 2 && re == 0 || engine == 1 && re == 1
+        continue
+      endif
+      let &regexpengine = engine
+      try
+        let l = matchlist(text, pat)
+      catch
+        call assert_report('Error ' . engine . ': pat: \"' . pat .
+		    \ '\", text: \"' . text .
+		    \ '\", caused an exception: \"' . v:exception . '\"')
+      endtry
+      " check the match itself
+      if len(l) == 0 && len(t) > matchidx
+        call assert_report('Error ' . engine . ': pat: \"' . pat .
+		    \ '\", text: \"' . text .
+		    \ '\", did not match, expected: \"' . t[matchidx] . '\"')
+      elseif len(l) > 0 && len(t) == matchidx
+        call assert_report('Error ' . engine . ': pat: \"' . pat .
+		    \ '\", text: \"' . text . '\", match: \"' . l[0] .
+		    \ '\", expected no match')
+      elseif len(t) > matchidx && l[0] != t[matchidx]
+        call assert_report('Error ' . engine . ': pat: \"' . pat .
+		    \ '\", text: \"' . text . '\", match: \"' . l[0] .
+		    \ '\", expected: \"' . t[matchidx] . '\"')
+      else
+        " Test passed
+      endif
+      if len(l) > 0
+        " check all the nine submatches
+        for i in range(1, 9)
+          if len(t) <= matchidx + i
+            let e = ''
+          else
+            let e = t[matchidx + i]
+          endif
+          if l[i] != e
+            call assert_report('Error ' . engine . ': pat: \"' . pat .
+                  \ '\", text: \"' . text . '\", submatch ' . i .
+                  \ ': \"' . l[i] . '\", expected: \"' . e . '\"')
+          endif
+        endfor
+        unlet i
+      endif
+    endfor
+  endfor
+  set regexpengine&
+endfunc
+
+" check that 'ambiwidth' does not change the meaning of \p
+func Test_ambiwidth()
+  set regexpengine=1 ambiwidth=single
+  call assert_equal(0, match("\u00EC", '\p'))
+  set regexpengine=1 ambiwidth=double
+  call assert_equal(0, match("\u00EC", '\p'))
+  set regexpengine=2 ambiwidth=single
+  call assert_equal(0, match("\u00EC", '\p'))
+  set regexpengine=2 ambiwidth=double
+  call assert_equal(0, match("\u00EC", '\p'))
+  set regexpengine& ambiwidth&
+endfunc
+
+" vim: shiftwidth=2 sts=2 expandtab